5-JCQ 1055

Confirmatory Factor Analysis of the Finnish Job Content Questionnaire (JCQ) in 590 Professional Musicians

1Orton Research Institute, Orton Foundation, Helsinki, Finland

2Department of Physical and Rehabilitation Medicine, Turku University Hospital and University of Turku, Turku, Finland

Correspondence to

Martti Vastamäki, Orton Research Institute, Orton Foundation, Tenholantie 10, 00280 Helsinki, Finland

Tel: +358-400-824-316

E-mail: martti.vastamaki@invalidisaatio.fi

Received: Mar 23, 2017

Accepted: Jun 9, 2017

Abstract

Background: Poorly functioning work environments may lead to dissatisfaction for the employees and financial loss for the employers. The Job Content Questionnaire (JCQ) was designed to measure social and psychological characteristics of work environments.

Objective: To investigate the factor construct of the Finnish 14-item version of JCQ when applied to professional orchestra musicians.

Methods: In a cross-sectional survey, the questionnaire was sent by mail to 1550 orchestra musicians and students. 630 responses were received. Full data were available for 590 respondents (response rate 38%).The questionnaire also contained questions on demographics, job satisfaction, health status, health behaviors, and intensity of playing music. Confirmatory factor analysis of the 2-factor model of JCQ was conducted.

Results: Of the 5 estimates, JCQ items in the “job demand” construct, the “conflicting demands” (question 5) explained most of the total variance in this construct (79%) demonstrating almost perfect correlation of 0.63. In the construct of “job control,” “repetitive work” (question 10) demonstrated a perfect correlation index of 0.84 and the items “little decision freedom” (question 14) and “allows own decisions” (question 6) showed substantial correlations of 0.77 and 0.65.

Conclusion: The 2-factor model of the Finnish 14-item version of JCQ proposed in this study fitted well into the observed data. The “conflicting demands,” “repetitive work,” “little decision freedom,” and “allows own decisions” items demonstrated the strongest correlations with latent factors suggesting that in a population similar to the studied one, especially these items should be taken into account when observed in the response of a population.

Keywords: Psychometrics; Factor analysis, statistical; Surveys and questionnaires; Job satisfaction; Music; Finland

Introduction

Work stress is recognized worldwide as a major risk to workers' health and the successful functioning of their organizations.1 Reduced motivation, deterioration in productivity and work safety, and depressive symptoms are common results of poorly functioning work environments leading to dissatisfaction for the employees as well as financial loss for the employers.1,2 Over 20 years ago, a self-administered instrument—the Job Content Questionnaire (JCQ)—was designed to measure social and psychological characteristics of work environments.3,4

The JCQ has gained great popularity and has undergone extensive study. The questionnaire has been translated into several languages containing a varying quantity of items.5-24 The modified Finnish version of the JCQ includes 12 or 14 items and has been used for many years.25 Studies on the JCQ have been conducted across a broad occupational spectrum including general populations7,20 as well as specific professions.5,6,8,9,11-13,15-19,23 Most research on the topic has concerned health-service workers.5,6,8,11,12,22,23 It has been suggested that the psychometric properties of JCQ may vary when applied to those in dissimilar occupations.10

The psychometric properties of the JCQ have not been studied amongst workers employed in any creative artistic environment. Such occupations may have their own explicit special features, and the expectation is that the impact of job strain may also affect artists differently than if they were members of the general population.26-31 The factor structures of questionnaires like JCQ may also be distinctive within this particular population and therefore, the JCQ may measure latent characteristics amongst musicians differing from those of other populations. A nationally representative sample of professional musicians may provide researchers and practitioners with a unique opportunity to investigate the measurement properties of the JCQ amongst people engaged in creative activity. Knowledge of JCQ psychometric properties may be useful when planning broader surveys and assessing occupational hazards in artistic work environments. The objective of this study was thus to investigate the factor construct of the Finnish 14-item version of the JCQ when applied to professional orchestra musicians.

Materials and Methods

The questionnaire was sent by mail to 1550 orchestra musicians and students (all 1000 Finnish orchestra musicians, all 500 musicians studying orchestra music in Sibelius Academy, and 50 retired orchestra musicians). The survey comprised 630 responses. Full data were available for 590 respondents (response rate 38%).The questionnaire also contained questions on demographics, job satisfaction, health status, health behaviors, and intensity of playing music. Their job status was categorized into “studying,” “working,” or “retired.” Perceived work ability was defined as a score on an 11-point numeric rating scale (NRS) from ‘0’ meaning “working is impossible” to ‘10’ indicating “the best work ability compared to the best level during a lifetime.” Perceived general health was assessed by a similar 11-point NRS ranging from ‘0’ meaning “the worst possible health” to ‘10’ indicating “the best possible health.” Respondent age was reported in full years on the day of response to the questionnaire without rounding.

Job Strain

According to the Karasek model, job strain was understood as a conflict between job demand and job control.3 In this study, it was assessed by using a 14-item Finnish adaptation of the JCQ. The first five items 1–5 described psychological “job demand;” the last items, 6–14, described “job control” (Fig 1). Responses were scored as ‘1’ meaning “completely agree,” ‘2’ “agree,” ‘3’ “cannot say,” ‘4’ “disagree,” and ‘5’ indicating “completely disagree.” The total score for job demand was calculated as a mean of the scores from items 1–5. When more than two responses were missing, the total score for items 1–5 was also considered missing. The total score for job control was calculated as the mean of individual scores from items 6–14. When more than five responses were missing, the total score for items 6–14 was also considered missing. The presence of job strain for each respondent was dichotomized as “yes” or “no.” Job strain was considered present if a job-demand score was above and a job-control score was below the sample's median value.

5-Fig%201-1055.jpg

Figure 1: Confirmatory factor analysis of 14-item Job Content Questionnaire (JCQ).

JCQ items—Q1: Work fast; Q2: Work hard; Q3: Excessive work; Q4: Insufficient time; Q5: Conflicting demands; Q6: Allows own decisions; Q7: Requires creativity; Q8: Learn new things; Q9: Repetitive work; Q10: Opinions influential; Q11: High skill level; Q12: Variety; Q13: Develop own abilities; Q14: Little decision freedom.

Circles represent unobserved and rectangles observed variables. ‘e’ variables represent a measurement error associated with the observed variable (variance that is not predicted by the latent factor). Single-headed arrows represent strength of correlation between two variables while double-headed arrows strength of correlation between two covariant variables.

Confirmatory Factor Analysis (CFA)

Estimating the model

The estimation procedure used the maximum likelihood method considering covariances supplied as input as being unbiased. For simplicity, the estimates were reported in standardized form as correlation coefficients. A correlation of <0.2 was considered “poor,” 0.21–0.4 “fair,” 0.41–0.6 “moderate,” 0.61–0.8 “substantial,” and >0.8 “perfect.”

Testing the model goodness of fit

In order to assess how well the model matched the observed data, the root mean square error of approximation (RMSEA) was used. First, the model fit was tested assuming there were no covariances between unique factors. After that, the modification indices suggested by the software were used to add covariance between factors (double-headed arrows in Fig 1) one at a time, each time testing the RMSEA closeness to the value of <0.05, or at least <0.08—the threshold for accepting the model fit.32,33 Every insertion was considered plausible if it made logical sense and did not violate the assumption that the common and the unique factors are uncorrelated. After achieving the RMSEA value of <0.05, no further covariances were imputed and the goodness of fit was assessed by χ2 test. As the sample was relatively small considering the requirements of CFA, in an attempt to reduce dependence on sample size, the choice was the relative (or “normed”) χ2 test. Relative χ2 is a χ2 estimate divided by the degrees of freedom. A relative χ2 value <5.0 was considered an indication of a good fit.34

All analyses were conducted using IBM® SPSS® Statistics for Windows®, ver 22.0 (IBM Corp. Released 2013, Armonk, NY, USA); IBM® SPSS® Amos™, ver 23.0 (IBM® Corp. Released 2013, PA, USA); and Stata/IC Statistical Software, release 14 (StataCorp LP, TX, USA).

Results

Descriptive Statistics

The mean age of 590 respondents was 37.5 (SD 12.8) years for the 338 (57%) men and 252 (43%) women, of whom 65% were working, 31% studying, and 4% retired. The median work history was 14 (IQR 5 to 25) years; 40% played violin or viola. Their perceived work ability was generally good, with a mean score of 8.1 (SD 1.9) of 10 points. Their general health they rated as a mean of 7.6 (SD 1.6) points. Job satisfaction was high with a median of 8 (IQR 7 to 9); the satisfaction with their salaries was low with a median of 3 (IQR 2 to 5). Of the respondents, 125 (21%) reported an elevated level of job strain, based on their responses to the JCQ.

Confirmatory Factor Analysis (CFA)

The model for describing the structure of the 14-item JCQ was built on two factors—“job psychological demand” and “job control,” as explained above. The first CFA run, conducted without any covariances between the single 14 JCQ items, produced a model fit that appeared to be inappropriate, with an RMSEA above 0.08. Using modification indices suggested by the software, covariances were imputed one at a time until the RMSEA decreased to 0.067 (90% CI 0.057 to 0.077), showing an acceptable fit as being below 0.08. At this point, the relative χ2 value was 3.5 (below the cut-off point of 5.0) with 58 degrees of freedom.4 In other words, the model presented in Figure 1 demonstrated good ability to describe the data from the study sample.

Of the five JCQ items in the “job demand” construct, the “conflicting demands” (question 5) explained most of the total variance in this construct (79%), demonstrating almost perfect correlation. A respective estimate of “excessive work” (question 3) showed a substantial correlation of 0.66. Other items of this construct indicated moderate correlations of 0.42 to 0.49.

In the construct of “job control,” “repetitive work” (question 10) demonstrated a perfect correlation index of 0.84 and the items “little decision freedom” (question 14) and “allows own decisions” (question 6) showed substantial correlations of a respective 0.77 and 0.65. Items 7 (“requires creativity”), 12 (“variety”), and 13 (“develop own abilities”) showed moderate correlations, item 8 (“learn new things”) fair, and items 9 (“repetitive work”) and 11 (“high skill level”) indicated poor correlations.

Discussion

In this cross-sectional study on 590 professional orchestra musicians investigating the factor construct of the Finnish version of 14-item JCQ, “conflicting demands” (question 5) explained most of the total variance in the “job demand” construct, whereas “repetitive work” (question 10) explained most of the total variance in the construct of “job control.”

This was the first study on the topic conducted in Finland among professional musicians—a profession that may differ from the general population because of its creative characteristics. Few studies have employed CFA to investigate the structure of the JCQ. The study conducted by Idrovo, et al, on a Mexican 8-item modified JCQ reported almost equal correlations for all eight items included in the proposed 3-factor model.35 Aboa-E´boule, et al, included nine items from the JCQ in a broader inventory considering all nine JCQ items related to an “extrinsic effort” latent factor.36 In that study, the strongest correlation (0.6) between unique and common variables was for a “work fast” item. The study by Żołnierczyk-Zreda, et al, tested the 4-factor model of a 52-item Polish version of the JCQ, reporting moderate or substantial correlations for almost every item.24 There is no evidence-based explanation for the strong correlation that we found here between “conflicting demands” and “repetitive work” and latent traits. We can only speculate that at least, in part, this finding can be explained by the specific features of the artistic profession here under study. For example, the discrepancy between poor satisfaction with salary and high work satisfaction that emerged in this sample could hardly be expected in the majority of occupations.

The main weakness of the study was low response rate (38%). This may affect the generalizability of the results across the entire field of professional musicians. Generalizability may also be affected by the differences between JCQ translations used in this and previous research.

Further research on the factor construct of the JCQ in diverse populations is necessary. Additional to factor structure, other psychometric properties of the JCQ should be investigated as well, for instance by Rasch analysis or item response theory.

In conclusion, the 2-factor model of the Finnish 14-item version of the JCQ proposed in this study fitted well the data obtained. The “conflicting demands,” “repetitive work,” “excessive work,” “little decision freedom,” and “allows own decisions” items demonstrated the strongest correlations with latent factors suggesting that in a population similar to this one, especially these items should be taken into account.

Acknowledgements

We thank Carol Norris, PhD, for language revision.

Conflicts of Interest: None declared.

Financial support: Finnish Social Insurance Institution, grant 9752

References

  1. Theorell T, Hammarstrom A, Aronsson G, et al. A systematic review including meta-analysis of work environment and depressive symptoms. BMC Public Health 2015;15:738.
  2. Leka S, Griffiths A, Cox T. Work Organization & stress. Protecting Workers' Health Series No. 3, WHO; France 2005.
  3. Karasek R, Brisson C, Kawakami N, et al. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol 1998;3:322-55.
  4. Karasek R, Choi B, Ostergren PO, et al. Testing two methods to create comparable scale scores between the Job Content Questionnaire (JCQ) and JCQ-like questionnaires in the European JACE Study. Int J Behav Med 2007;14:189-201.
  5. Alexopoulos EC, Argyriou E, Bourna V, et al. Reliability and Validity of the Greek Version of the Job Content Questionnaire in Greek Health Care Workers. Saf Health Work 2015;6:233-9.
  6. Amin NA, Quek KF, Oxley JA, et al. Validity and Reliability of Malay Version of the Job Content Questionnaire among Public Hospital Female Nurses in Malaysia. Int J Occup Environ Med 2015;6:232-42.
  7. Cheng Y, Luh WM, Guo YL. Reliability and validity of the Chinese version of the Job Content Questionnaire in Taiwanese workers. Int J Behav Med 2003;10:15-30.
  8. Chien TW, Lai WP, Wang HY, et al. Applying the revised Chinese Job Content Questionnaire to assess psychosocial work conditions among Taiwan's hospital workers. BMC Public Health 2011;11:478-86.
  9. Choi B, Kurowski A, Bond M, et al. Occupation-differential construct validity of the Job Content Questionnaire (JCQ) psychological job demands scale with physical job demands items: a mixed methods research. Ergonomics 2012;55:425-39.
  10. Choi B, Ko S, Dobson M, et al. Short-term test-retest reliability of the Job Content Questionnaire and Effort-Reward Imbalance Questionnaire items and scales among professional firefighters. Ergonomics 2014;57:897-911.
  11. Choobineh A, Ghaem H, Ahmedinejad P. Validity and reliability of the Persian (Farsi) version of the Job Content Questionnaire: a study among hospital nurses. East Mediterr Health J 2011;17:335-41.
  12. Eum KD, Li J, Jhun HJ, et al. Psychometric properties of the Korean version of the job content questionnaire: data from health care workers. Int Arch Occup Environ Health 2007;80:497-504.
  13. Hadi AA, Naing NN, Daud A, et al. Reliability and construct validity of the Malay version of the Job Content Questionnaire (JCQ) among secondary school teachers in Kota Bharu, Kelantan, Malaysia. Southeast Asian J Trop Med Public Health 2006;37:1254-9.
  14. Hoang TG, Corbiere M, Negrini A, et al. Validation of the Karasek-Job Content Questionnaire to measure job strain in Vietnam. Psychological reports 2013;113:363-79.
  15. Kawakami N, Fujigaki Y. Reliability and validity of the Japanese version of Job Content Questionnaire: replication and extension in computer company employees. Ind Health 1996;34:295-306.
  16. Kawakami N, Kobayashi F, Araki S, et al. Assessment of job stress dimensions based on the job demands- control model of employees of telecommunication and electric power companies in Japan: reliability and validity of the Japanese version of the Job Content Questionnaire. Int j behav med 1995;2:358-75.
  17. Li W, Zhang JQ, Sun J, et al. Reliability and validity of Job Content Questionnaire in Chinese petrochemical employees. Psychol Rep 2007;100:35-46.
  18. Maizura H, Masilamani R, Aris T. Reliability (internal consistency) of the job content questionnaire on job stress among office workers of a multinational company in Kuala Lumpur. Asia Pac J Public Health 2009;21:216-22.
  19. Nehzat F, Huda BZ, Tajuddin SH. Reliability and validity of job content questionnaire for university research laboratory staff in Malaysia. Southeast Asian J Trop Med Public Health 2014;45:481-9.
  20. Phakthongsuk P. Construct validity of the Thai version of the job content questionnaire in a large population of heterogeneous occupations. J Med Assoc Thai 2009;92:564-72.
  21. Phakthongsuk P, Apakupakul N. Psychometric properties of the Thai version of the 22-item and 45-item Karasek job content questionnaire. Int J Occup Med Environ Health 2008;21:331-344.
  22. Poanta LI, Zdrenghea D, Albu A. Psychometric evaluation of Romanian version of Job Content Questionnaire in physicians. Rom J Intern Med 2006;44:183-99.
  23. Tabatabaee Jabali SM, Ghaffari M, Pournik O, et al. Reliability and validity of Persian version of job content questionnaire in health care workers in Iran. Int J Occup Environ Med 2013;4:96-101.
  24. Zolnierczyk-Zreda D, Bedynska S. Psychometric properties of the Polish version of Karasek's Job Content Questionnaire. Int J Occup Saf Ergon 2014;20:583-93.
  25. Kouvonen A, Kivimäki M, Elovainio M, et al. Low organisational justice and heavy drinking: a prospective cohort study. Occupat Environm Med 2008;65:44-50.
  26. Bartel LR, Thompson EG. Coping With Perfortllance Stress: A Study Of Professional Orchestral Musicians In Canada. Quarterly J Music Teaching Learning 2014;4:70-8.
  27. Dommerholt J. Performing arts medicine - instrumentalist musicians part I - general considerations. J Bodyw Mov Ther 2009;13:311-9.
  28. Hoppmann RA. Instrumental musicians' hazards. Occup Med 2001;16:619-31,iv-v.
  29. Kenny DT, Davis P, Oates J. Music performance anxiety and occupational stress amongst opera chorus artists and their relationship with state and trait anxiety and perfectionism. J Anxiety Disord 2004;18:757-77.
  30. Parasuraman S, Purohit YS. Distress and boredom among orchestra musicians: the two faces of stress. J Occup Health Psychol 2000;5:74-83.
  31. Pereira ÉF, Kothe F, de Souza Bleyer FT, et al. Work-related stress and musculoskeletal complaints of orchestra musicians. Revista Dor 2014;15:112-6.
  32. Browne MW, Cudec R. Alternative ways of assessing model fit in testing structural equation models. In: Bollen KA, Long JS, eds. Testing structural equation models. New York: SAGE Focus Editions. 1993;154:136-162.
  33. Steiger JH. Structural Model Evaluation and Modification: An Interval Estimation Approach. Multivariate Behav Res 1990;25:173-180.
  34. Schumacker RE, Lomax RG. A beginner's guide to structural equation modeling. 2nd ed. Mahwah, NJ: Lawrence Erlbaum Associates, 2004.
  35. Idrovo AJ, Camacho-Avila A, Garcia-Rivas J, et al. Social capital at work: psychometric analysis of a short scale in Spanish among Mexican health workers. Rev Bras Epidemiol 2012;15:536-47.
  36. Aboa-Eboule C, Brisson C, Blanchette C, et al. Effort-reward imbalance at work and psychological distress: a validation study of post-myocardial infarction patients. Psychosom Med 2011;73: 448-55.

TAKE-HOME MESSAGE

  • Work stress reduces motivation, causes deterioration in productivity and work safety leading to dissatisfaction for the employees and financial loss for the employers.
  • The Job Content Questionnaire (JCQ) is a good tool to measure social and psychological characteristics of work environments.
  • Psychometric properties of the JCQ may vary when applied to those in dissimilar occupations and have not been studied amongst workers employed in any creative artistic environment.
  • The “conflicting demands,” “repetitive work,” “little decision freedom,” and “allows own decisions” items demonstrated the strongest correlations with latent factors suggesting that amongst workers employed in any creative artistic environment similar to the studied one, especially these items should be taken into account.-

Cite this article as: Vastamäki H, Vastamäki M, Laimi K, Saltychev M. Confirmatory factor analysis of the Finnish job content questionnaire (JCQ) in 590 professional musicians. Int J Occup Environ Med 2017;8:-180.




 pISSN: 2008-6520
 eISSN: 2008-6814

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License